The Creep function moves a stage until a predefined load variation (Load Target (relative to initial load)) is reached and maintains this loading through real-time closed-loop PID adjustment of the stage position. This function can be used, for example, to study the deformation of a sample under constant loading. Note that tuning a PID requires a reasonable amount of closed-loop system understanding. A set of PID parameters optimized for a particular experimental configuration will probably be inappropriate for a different configuration and could even create system instabilities (oscillations) that can result in damage to the system.
Parameters:
STAGE AXIS - On multi-axis systems, the user must specify which axis will perform the “Creep”.
DIRECTION - Specify if the stage has to start moving in the positive or negative direction. With a positive displacement, a vertical stage will move down (compression), a horizontal stage will move right, and a rotation stage will move clockwise. Enter a negative value for stage displacements in opposite directions.
STAGE VELOCITY, MM/S - Specify the speed at which the stage will move. Refer to the specifications of your MachOne? model to learn about the velocity range allowed. In MachOne Motion, the speed is always specified in absolute value terms.
RELATIVE LOAD TARGET, GF OR N - Specify the relative amount of load to be applied and maintained on the sample. When a value is entered, the stage will initiate its movement until a change in load of that magnitude is reached. The value for Relative Load Target is specified in absolute value terms.
CREEP TIME, SEC - This is the time period during which the relative loading will be maintained. Note that after this period, the final stage position will be maintained.
P GAIN - This is the Proportional gain of the PID control loop. This is the main parameter that drives the feedback loop. Our experiments revealed that a good starting point for the optimization of this parameter is given by P = 0.001 (and I = 0), then raising P in logarithmic increments (0.01, 0.1, 1) until a satisfactory amplitude in reached. The higher the P value, the faster the error will be corrected, however if the P is made too large, the error can go negative (actual load>desired load) which may start a process where the stage starts to oscillate between speeding up and slowing down. This oscillation can result in the motor control system going unstable. A simple way to eliminate this instability would be to decrease P until the oscillation stops.
I GAIN - This is the Integral gain of the PID control loop. Integral control sums the error from each position update then multiplies the sum by the Integral gain, I. Our experiments revealed that for most creep tests, best results are achieved with I = 0. If a residual error exists while holding a load, the I value can also be incrementally increased until the error disappears.
